Technology and Digital Transformation

Introduction: From Information Technology to Enterprise Reinvention

Technology has moved from being a support function to becoming a central determinant of competitive advantage, operating-model design, customer experience, and organizational resilience. Digital transformation is not simply the installation of new software. It is the deliberate redesign of how an organization creates, delivers, and captures value through data, digital platforms, automation, connectivity, and organizational change.

For managers, the central question is not “Which technology should we buy?” but “Which strategic problem are we solving, and what combination of technology, processes, capabilities, governance, and leadership is required?” A technically sophisticated solution can fail if it is disconnected from business priorities, poorly adopted, insecure, or economically unjustified. Conversely, relatively simple technologies can produce substantial value when they remove friction from a strategically important process.

Digital transformation is therefore a socio-technical undertaking. Its outcomes depend on the interaction of technology with people, structures, incentives, routines, suppliers, regulators, and customers. This chapter examines the technologies, platforms, automation approaches, digital operating models, and leadership practices required to manage transformation at enterprise scale.

Learning Objectives

By the end of this chapter, the reader should be able to:

1. Distinguish digitization, digitalization, and digital transformation.
2. Assess enterprise technologies and platforms in relation to strategic objectives.
3. Explain cloud, data, artificial intelligence, cybersecurity, and integration architectures.
4. Evaluate automation opportunities using economic, operational, and ethical criteria.
5. Design and critique digital operating models and transformation road maps.
6. Apply governance, risk, and change-management principles to technology-enabled transformation.
7. Lead a transformation portfolio using measurable outcomes rather than technology adoption alone.

1. Core Concepts and Strategic Logic

1.1 Digitization, digitalization, and transformation

These terms describe different levels of change:

The distinction matters because organizations often report successful digitization while leaving underlying processes unchanged. A paper form converted into a poorly designed online form may reduce printing but not improve customer value. Transformation requires changes in decision rights, processes, capabilities, metrics, and sometimes revenue logic.

1.2 The transformation value chain

A useful managerial chain is:

Strategic intent → customer or operational problem → redesigned process → enabling technology → adoption and governance → measurable outcome.

Technology is an enabler, not the value outcome. Outcomes may include reduced cycle time, lower cost-to-serve, improved quality, higher retention, faster innovation, better compliance, or new revenue. A business case should identify a baseline, target, time horizon, investment, risks, and ownership.

1.3 Dynamic capabilities

The dynamic-capabilities perspective explains why digital advantage is difficult to sustain. Organizations must:

• Sense changes in customer expectations, technology, regulation, and competitors.
• Seize opportunities by mobilizing investment, talent, partnerships, and products.
• Transform assets, routines, and structures so that the organization can continue adapting.

Technology can strengthen these capabilities, but it cannot substitute for managerial judgment. A firm with advanced analytics may still fail to act on insights because authority is centralized, incentives are misaligned, or employees do not trust the data.

2. Enterprise Technology Landscape

2.1 Cloud computing

Cloud computing provides on-demand access to computing resources, storage, databases, platforms, and software, usually with elastic capacity and usage-based pricing. Common service models are:

• Infrastructure as a Service (IaaS): virtual machines, networks, and storage.
• Platform as a Service (PaaS): managed environments for developing and deploying applications.
• Software as a Service (SaaS): complete applications delivered to users.
• Function as a Service (FaaS): event-triggered execution of small units of code.

Deployment choices include public, private, hybrid, and multi-cloud arrangements. Cloud decisions should consider workload characteristics, data sovereignty, latency, resilience, integration, skills, vendor concentration, and total cost of ownership. “Cloud first” is not equivalent to “cloud only.” Moving an inefficient process to the cloud may simply reproduce inefficiency at a different cost structure.

Cloud economics also require attention to variable consumption. Costs may increase unexpectedly through data transfer, excessive storage, duplicated environments, or poorly governed computing demand. FinOps practices connect engineering decisions with financial accountability by assigning cost ownership, monitoring utilization, and optimizing resource consumption.

2.2 Enterprise platforms and ecosystems

Enterprise platforms provide shared capabilities on which multiple products, functions, or partners can operate. Examples include enterprise resource planning, customer relationship management, supply-chain platforms, digital commerce, data platforms, identity services, and collaboration suites.

Platforms create value through standardization, reuse, network effects, and faster product development. They also create risks. A platform can impose rigid processes, concentrate operational dependency, increase switching costs, and spread a failure across many business units. Platform selection should therefore assess:

1. Strategic fit and process coverage.
2. Extensibility and integration capability.
3. Data ownership and portability.
4. Security, resilience, and compliance.
5. Vendor viability and ecosystem quality.
6. Implementation complexity and total cost.
7. User experience and adoption requirements.

An enterprise architecture should distinguish systems of record, systems of differentiation, and systems of innovation. Systems of record maintain authoritative transactions; systems of differentiation support distinctive capabilities; systems of innovation enable experimentation. This classification helps managers avoid applying the same governance and delivery methods to every system.

2.3 Data and analytics architecture

Data is valuable only when it is accessible, reliable, interpretable, protected, and connected to decisions. Important architectural concepts include:

• Data warehouse: curated, structured data designed primarily for reporting and analysis.
• Data lake: large-scale storage for raw or varied data, including unstructured data.
• Lakehouse: an architecture seeking to combine the flexibility of a lake with the governance and analytical performance of a warehouse.
• Data mesh: a decentralized approach in which domain teams own data products while sharing common standards and infrastructure.
• Master data management: processes for maintaining consistent core entities such as customers, products, suppliers, and locations.

Data governance defines ownership, quality standards, access rules, retention, lineage, privacy, and accountability. A dashboard can create false confidence if definitions differ across departments. For example, “active customer” might mean a customer who purchased in the last 30 days in one system and the last 12 months in another.

A practical data-quality model evaluates accuracy, completeness, consistency, timeliness, uniqueness, and validity. Data quality should be managed at the point of capture where possible, rather than corrected indefinitely downstream.

2.4 Artificial intelligence and machine learning

Artificial intelligence includes systems that perform tasks associated with perception, reasoning, generation, prediction, or decision support. Machine learning systems infer patterns from data rather than relying solely on explicitly programmed rules. Generative AI produces text, images, code, audio, or other content from learned representations.

Business applications include demand forecasting, fraud detection, predictive maintenance, service assistance, document processing, recommendation, quality inspection, and knowledge retrieval. However, a model’s technical performance is only one part of business value. Managers must also evaluate interpretability, data rights, bias, security, human oversight, process integration, and the consequences of error.

A classification model can be evaluated using a confusion matrix. Let true positives be TP, true negatives TN, false positives FP, and false negatives FN.

Accuracy is appropriate only when the costs of error are reasonably balanced and class proportions are not misleading. For an imbalanced fraud-detection problem, precision and recall may be more informative:

Precision measures the proportion of positive predictions that are correct. Recall measures the proportion of actual positive cases detected. The formulas assume binary classification and nonzero denominators. A bank deciding whether to investigate transactions may prioritize recall, while a small compliance team with limited investigation capacity may place greater emphasis on precision. The correct threshold is therefore a business and risk decision, not merely a technical one.

Generative AI introduces additional risks, including fabricated outputs, confidential-data leakage, prompt injection, intellectual-property disputes, and automation bias. Appropriate controls include approved use cases, access restrictions, retrieval from authoritative sources, output validation, logging, human review for material decisions, and model-risk monitoring.

2.5 Internet of Things, edge computing, and digital twins

Internet of Things systems connect physical objects, sensors, machines, and environments to networks and applications. Edge computing processes data near the source when latency, bandwidth, reliability, or privacy makes centralized processing unsuitable. A digital twin is a digital representation of a physical asset, process, or system that is updated with operational data and used for monitoring, simulation, or optimization.

These technologies are important in manufacturing, logistics, energy, healthcare, and infrastructure. Their value depends on reliable sensors, interoperable standards, cybersecurity, data ownership, and the ability to connect insight to operational action. Collecting more data does not automatically improve decisions.

2.6 Cybersecurity and resilience

Cybersecurity is a business risk discipline, not merely an information-technology concern. A useful foundation is the protection of confidentiality, integrity, and availability. Contemporary programs also emphasize identity, supply-chain risk, secure development, incident response, recovery, and resilience.

Zero-trust architecture treats access as continuously verified rather than automatically trusted because a user or device is inside a network boundary. Controls include strong identity management, least privilege, segmentation, encryption, endpoint protection, monitoring, backups, and tested recovery procedures.

Risk can be represented conceptually as a function of threat, vulnerability, exposure, and impact. This is not a universal numerical equation; it is a decision model. Quantitative estimates should be used only when assumptions and data quality justify them. Boards should understand not only the probability of an incident but also operational interruption, regulatory consequences, customer harm, litigation, and reputational effects.

3. Automation and Process Redesign

3.1 Choosing what to automate

Automation ranges from simple rules and workflow to robotic process automation, intelligent document processing, machine learning, and autonomous decision systems. The correct starting point is process analysis, not tool selection.

A process is a strong automation candidate when it is frequent, rules-based, stable, digitally observable, and sufficiently standardized. It is a weak candidate when it depends on ambiguous judgment, poorly defined exceptions, unreliable data, or frequent policy changes.

Managers should ask:

• Is the process valuable enough to justify change?
• Can unnecessary steps be removed before automation?
• Are inputs and outputs defined?
• What is the error cost?
• What human judgment must remain?
• How will exceptions be escalated?
• Can the process be monitored and reversed?

3.2 Economic evaluation

A simple annual return-on-investment measure is:

ROI is a ratio, usually expressed as a percentage. Benefits and costs should be measured over the same period and stated consistently in nominal or real monetary terms. The formula is useful for screening but ignores timing, risk, and residual value. For larger investments, discounted cash flow is more appropriate:

Here, CF_t is net cash flow in period t, r is the discount rate, and n is the number of periods. The formula assumes cash flows are estimated at defined time intervals and that the discount rate reflects risk and opportunity cost. A positive NPV indicates value creation relative to the chosen discount rate, subject to the quality of assumptions.

Worked Example: Automation Investment

A service center considers an automation project requiring an initial investment of $240,000. It expects annual benefits of $150,000 and annual operating costs of $30,000 for three years. The net annual benefit is therefore $120,000.

Using a discount rate of 10%:

NPV = −240,000 + 120,000 ÷ 1.10 + 120,000 ÷ 1.10² + 120,000 ÷ 1.10³

NPV ≈ −240,000 + 109,091 + 99,174 + 90,158 = −$41,577.

Although the simple undiscounted net benefit over three years is $120,000, the NPV is negative at a 10% discount rate. The project may still be justified by strategic, compliance, or capacity benefits, but those benefits must be made explicit rather than hidden in an inflated financial estimate.

3.3 Human–automation complementarity

Automation should be designed around complementarity rather than replacement alone. Machines are often strong at scale, consistency, pattern recognition, and repetitive processing. Humans remain essential for contextual interpretation, empathy, accountability, negotiation, ethical judgment, and handling novel situations.

A mature operating design specifies automation boundaries, human approval points, exception paths, audit trails, and performance monitoring. It also addresses job redesign, reskilling, psychological safety, and workforce representation. Poorly managed automation can reduce autonomy, intensify work, and create hidden surveillance even when headline productivity improves.

4. Digital Operating Models

A digital operating model describes how an organization structures people, processes, technology, governance, funding, and performance management to deliver digital outcomes.

4.1 Common models

Many enterprises use a hybrid arrangement. Shared platforms, security, architecture, data standards, and core infrastructure may be centralized, while cross-functional product teams own customer journeys or operational capabilities.

4.2 Product teams and agile delivery

A product team typically includes product management, design, engineering, data, operations, and relevant domain expertise. It owns a product or journey over time rather than delivering a project and disbanding. Agile methods support incremental delivery, feedback, prioritization, and adaptation, but agile ceremonies alone do not create agility. Structural barriers include annual funding cycles, fragmented ownership, excessive approval layers, and metrics that reward output rather than outcomes.

DevOps and platform engineering connect software development with reliable operations. Practices such as automated testing, continuous integration, continuous delivery, observability, infrastructure as code, and controlled release reduce the time and risk of change. They require investment in architecture, skills, controls, and culture.

4.3 Measuring transformation

Technology metrics such as uptime, deployment frequency, and system performance are necessary but insufficient. A balanced scorecard may include:

• Customer: satisfaction, task completion, retention, complaints.
• Operational: cycle time, error rate, straight-through processing, productivity.
• Financial: revenue, margin, cost-to-serve, cash impact, NPV.
• People: adoption, capability development, engagement, attrition.
• Risk: control effectiveness, incidents, recovery time, compliance.
• Innovation: experiment velocity, learning rate, successful product releases.

A useful distinction is between output metrics and outcome metrics. “Number of features released” is an output. “Reduction in customer abandonment” is an outcome. Leaders should link investment decisions to outcomes while retaining technical health indicators to prevent short-term optimization.

5. Transformation Governance and Leadership

5.1 Portfolio governance

Transformation should be managed as a portfolio rather than a collection of unrelated projects. Portfolio governance evaluates strategic alignment, dependencies, capacity, risk, benefits, and sequencing. A transformation office can coordinate the portfolio, but it should not become a reporting bureaucracy detached from delivery.

Decision rights should be explicit. Boards oversee strategic risk and investment. Executives set priorities and resolve cross-functional conflicts. Product and process owners are accountable for outcomes. Technology and security leaders establish enabling standards and challenge unacceptable risk. Employees and customers provide evidence about usability and consequences.

5.2 Architecture and technical debt

Technical debt is the future cost created by expedient technical decisions, obsolete platforms, duplicated data, fragile integrations, or inadequate documentation. Debt is not always bad; deliberate shortcuts can be rational when learning speed is more valuable than immediate elegance. Unmanaged debt becomes dangerous when it increases failure risk, slows change, or prevents compliance.

Architecture principles should address modularity, interoperability, security by design, resilience, observability, data portability, and lifecycle management. Standards should enable reuse without suppressing appropriate innovation.

5.3 Change leadership

Transformation changes identity, power, work practices, and resource allocation. Effective leadership therefore combines a compelling case for change with participation, transparency, experimentation, and visible sponsorship. Stakeholder analysis should identify who gains, who bears costs, whose work changes, and who can block progress.

A practical change sequence is:

1. Establish the strategic case and measurable ambition.
2. Map current capabilities, processes, data, technology, and constraints.
3. Prioritize a portfolio of high-value opportunities.
4. Design the target operating model and architecture.
5. Deliver in increments with users involved.
6. Measure adoption, outcomes, and unintended effects.
7. Scale what works and retire what does not.

6. Ethics, Regulation, and Responsible Technology

Digital transformation creates responsibilities beyond shareholder value. Privacy requires lawful, fair, and transparent handling of personal data; data minimization and purpose limitation reduce unnecessary exposure. Explainability is especially important where automated decisions affect employment, credit, healthcare, education, insurance, or access to services.

Responsible AI governance should include:

• documented purpose and accountability;
• assessment of data provenance and representativeness;
• testing for performance and disparate impact;
• human oversight proportionate to risk;
• security against manipulation and unauthorized disclosure;
• monitoring after deployment;
• mechanisms for appeal, correction, and incident response.

Regulatory obligations vary by jurisdiction and sector. Managers should not assume that compliance is achieved because a vendor labels a product “AI compliant.” Responsibility is shared across the provider, deployer, process owner, and senior leadership.

Digital inclusion is also a strategic and ethical issue. A digital channel may reduce cost for some users while excluding people with limited connectivity, disability, language barriers, low digital literacy, or distrust of institutions. Inclusive design, alternative channels, accessibility standards, and user research are therefore part of transformation quality.

7. Applied Case: Northstar Industrial Services

The following is an illustrative case.

Background

Northstar Industrial Services maintains equipment for commercial customers. Its revenue historically came from scheduled maintenance and emergency repairs. The company operated a centralized enterprise-resource-planning system, spreadsheets owned by regional offices, and a call center that manually scheduled technicians.

Problem

Customers experienced long waits for repairs, while technicians frequently arrived without the correct parts. Management considered purchasing an AI platform, but the underlying data was inconsistent: asset identifiers differed between regions, maintenance histories were incomplete, and sensor data was not connected to work orders. The company also faced a shortage of experienced technicians.

Analysis

A diagnostic team mapped the service journey and found that the largest delays occurred in three activities: identifying the installed asset, diagnosing likely failure, and locating parts. The team estimated that technology would create value only if master data, integration, technician workflows, and decision rights were redesigned together.

The proposed architecture included connected equipment sensors, an event-streaming layer, a governed asset master, a service-management platform, a mobile technician application, and an analytics model for failure-risk prioritization. The model would recommend inspection priority but would not automatically dispatch technicians without human review during the pilot.

Approach and solution

Northstar used a product operating model with a cross-functional service product team. It standardized asset identifiers, established data ownership with regional participation, and implemented APIs between the enterprise system, service platform, parts inventory, and mobile application. A limited pilot covered one equipment category and two regions.

The team measured first-time-fix rate, mean time to repair, parts availability, customer wait time, technician adoption, model precision, and safety incidents. Training included not only application use but also how recommendations were generated and when technicians should override them. A governance board reviewed model drift, customer complaints, and workforce effects monthly.

Outcome

After six months, the pilot reduced average scheduling delay and improved first-time-fix performance. The financial benefit was lower than the original business case because sensor installation costs were underestimated. However, the pilot generated reliable evidence, improved data quality, and revealed that some equipment types were unsuitable for predictive maintenance. Northstar expanded the solution selectively rather than deploying it universally.

Learning points

1. The business problem, not the AI label, defined the transformation.
2. Data governance and process redesign were prerequisites for analytics value.
3. A bounded pilot reduced technical, operational, and ethical risk.
4. Human judgment remained essential in safety-relevant decisions.
5. Outcome metrics and learning were more valuable than a simplistic technology roll-out target.

8. Contemporary Developments and Strategic Implications

Several developments are reshaping transformation. Generative AI is lowering the cost of content, coding, search, and interaction, but it is increasing the importance of information governance and verification. Industry cloud platforms combine sector-specific data models, workflows, and controls. Digital public infrastructure and interoperable identity systems are changing how organizations deliver services. Edge computing and 5G support more responsive industrial and field applications. Sustainability pressures are making energy consumption, hardware lifecycles, electronic waste, and carbon reporting material technology decisions.

The strategic implication is that technology portfolios must be managed for adaptability. Leaders should avoid both technological determinism—the belief that technology automatically creates progress—and technological conservatism—the belief that existing processes should be protected from digital challenge. The strongest organizations combine experimentation with disciplined governance, local learning with enterprise standards, and speed with resilience.

Best-Practice Checklist

Before approving a major technology-enabled transformation, management should be able to answer:

• What strategic or customer outcome is being improved?
• What is the baseline and how will improvement be measured?
• Which process, capability, and behavior must change?
• Who owns the outcome after implementation?
• What data, integration, security, and resilience requirements exist?
• What assumptions drive the business case?
• What risks could harm customers, employees, or society?
• What is the smallest credible pilot?
• How will learning influence the next investment decision?
• What will be retired, simplified, or stopped?

Chapter Summary

Digital transformation is the redesign of value creation and operating capability through technology, data, automation, and organizational change. Digitization converts information, digitalization improves existing processes, and transformation changes the business or operating model. Cloud platforms, enterprise systems, data architectures, artificial intelligence, connected devices, and cybersecurity form an interdependent technology landscape rather than a collection of isolated tools.

Automation should begin with process and value analysis. Financial measures such as ROI and NPV support decisions but must be supplemented by strategic, operational, risk, workforce, and ethical analysis. Digital operating models increasingly use cross-functional product teams, shared platforms, agile delivery, DevOps, and outcome-based measurement. Governance must address architecture, technical debt, data quality, security, privacy, responsible AI, resilience, and inclusion.

Successful transformation is led as a portfolio of measurable learning and outcomes. It requires executive sponsorship, clear decision rights, capable teams, user participation, disciplined experimentation, and willingness to retire ineffective initiatives. Technology creates advantage when it is integrated with distinctive organizational capabilities and governed according to the consequences of its use.

Key Terms

Digitization: Conversion of analogue information into digital form.

Digitalization: Use of digital technology to improve an existing activity or process.

Digital transformation: Enterprise-level redesign of value creation, delivery, or capture through digital capabilities.

Cloud computing: On-demand provision of computing resources over a network.

Platform: Shared technological and organizational foundation used by multiple products, functions, or participants.

Enterprise architecture: The structured relationship among strategy, processes, information, applications, and technology.

Data governance: Decision rights, standards, controls, and accountability for data.

Master data management: Management of consistent, authoritative records for core business entities.

Artificial intelligence: Systems that perform tasks involving perception, reasoning, generation, prediction, or decision support.

Machine learning: Methods in which systems infer patterns from data to perform tasks.

Generative AI: AI that produces new content such as text, images, code, or audio.

Automation: Use of technology to perform tasks or processes with reduced direct human intervention.

Robotic process automation: Software automation of structured, repetitive interactions with digital systems.

Digital operating model: The arrangement of people, processes, technology, governance, funding, and metrics used to deliver digital outcomes.

Product operating model: Continuous ownership of a product, service, or journey by a cross-functional team.

Technical debt: Future cost or risk created by expedient or obsolete technical decisions.

Zero trust: Security approach requiring continuous verification rather than assuming inherent trust based on network location.

Dynamic capabilities: Organizational abilities to sense, seize, and transform in response to change.

Responsible AI: Design and use of AI with attention to accountability, fairness, safety, transparency, privacy, and human oversight.

Digital twin: A data-connected digital representation of a physical asset, process, or system.

Review Questions

Short Answer Questions

1. Distinguish digitization, digitalization, and digital transformation.
2. What are the principal service models of cloud computing?
3. Why is data governance important to enterprise analytics?
4. Define technical debt and explain one risk it creates.
5. Why should automation projects include human exception handling?

Long Answer Questions

1. Critically evaluate the advantages and risks of using enterprise platforms in a multinational organization.
2. Explain how a product operating model differs from a traditional project-based model.
3. Discuss the managerial implications of generative AI for data governance, work design, and accountability.
4. Explain why cybersecurity should be treated as an enterprise-risk issue rather than solely an IT issue.
5. Develop a transformation-governance approach for an organization with centralized IT and highly autonomous business units.

Analytical / Application Questions

1. A firm proposes automating a process that is frequent but poorly standardized and dependent on expert judgment. Analyze whether it is a suitable candidate and recommend a staged approach.
2. Using the NPV formula in this chapter, evaluate an investment with an initial cost of $500,000 and net cash benefits of $220,000, $240,000, and $260,000 over three years at a 12% discount rate. State the limitations of your evaluation.
3. Design a responsible-AI control framework for a customer-credit decision system, including data, model, human oversight, monitoring, appeal, and cybersecurity controls.

Multiple Choice Questions with Answers

Question 1

Which statement best distinguishes digital transformation from digitization?

A. Digitization always requires artificial intelligence.

B. Digital transformation redesigns value creation or operating capability, whereas digitization converts analogue information into digital form.

C. Digital transformation is limited to replacing paper records.

D. Digitization changes an organization’s business model by definition.

Correct Answer: B. Digital transformation redesigns value creation or operating capability, whereas digitization converts analogue information into digital form.

Question 2

Which cloud service model provides a complete application to end users?

A. Infrastructure as a Service

B. Platform as a Service

C. Software as a Service

D. Function as a Service

Correct Answer: C. Software as a Service

Question 3

A fraud model identifies 80 true positives and 20 false positives. What is its precision?

A. 20%

B. 25%

C. 75%

D. 80%

Correct Answer: D. 80%

Question 4

Which condition generally makes a process a strong automation candidate?

A. Frequent execution and stable, rules-based activities

B. Unclear inputs and constantly changing exceptions

C. Dependence on undocumented expert intuition

D. High ambiguity with no measurable output

Correct Answer: A. Frequent execution and stable, rules-based activities

Question 5

What is the primary purpose of master data management?

A. To eliminate all unstructured data

B. To maintain consistent authoritative records for core entities

C. To replace cybersecurity controls

D. To ensure every department uses a different data definition

Correct Answer: B. To maintain consistent authoritative records for core entities

Question 6

Which measure is an outcome metric rather than an output metric?

A. Number of features released

B. Number of training sessions delivered

C. Reduction in customer abandonment

D. Number of project meetings held

Correct Answer: C. Reduction in customer abandonment

Question 7

What does a zero-trust security approach assume?

A. Internal network users are automatically trusted

B. Only external users require authentication

C. Access should be continuously verified and appropriately limited

D. Encryption is unnecessary inside the organization

Correct Answer: C. Access should be continuously verified and appropriately limited

Question 8

In the Northstar case, why was data standardization necessary?

A. To make sensor installation unnecessary

B. To create consistent asset identities across regions and systems

C. To remove human review from safety decisions

D. To guarantee that every equipment type would benefit from predictive maintenance

Correct Answer: B. To create consistent asset identities across regions and systems

Question 9

An investment has an initial cost of $100,000 and produces one net cash benefit of $120,000 after one year. At a 10% discount rate, what is its approximate NPV?

A. −$9,091

B. $0

C. $9,091

D. $20,000

Correct Answer: C. $9,091

Question 10

Which practice is most consistent with responsible AI governance?

A. Deploying a model without monitoring once accuracy is acceptable

B. Allowing a vendor to make all accountability decisions

C. Using human oversight, documented purpose, monitoring, and appeal mechanisms for high-impact decisions

D. Maximizing automation by removing exception paths

Correct Answer: C. Using human oversight, documented purpose, monitoring, and appeal mechanisms for high-impact decisions